Rapid Antigen tests are very sensitive at Ct < 30

Above which-too little live virus to detect

Ppl worry transmission could occur at Ct > 30 and thus missed by Ag test

Transmissibility is NOT BINARY
Possible at 34? Yes
Likely? No.

This graph helps place Cts in context!

1/
(If do not see the full graph above - please click on it)

Easy to forget Ct values are LOG scale

They place as much emphasis on an increase of 1000 viral particles as they do an increase of 1,000,000,000!!

2/
But when we change from a Ct scale to a scale that shows the number of viruses withouth "zooming in" on the low end, we see just how stark a different 20-25 is versus 30-35.

3/
Nearly all of the concern about rapid antigen test sensitivity is about Cts 30-35.

But why? For no great reason except that that that is where viral culture / transmissibility turns neg.

But It is SO much more important to ensure you're catching the Ct 15-28, than the 32

4/
We say "well, it IS possible to get a culturable sample at 32"...

Perhaps so!! But like transmissibility, culturability is on a HUGE continuum. Just because it's culturable and potentially transmissible doesn't mean it's very likely to transmit.

5/
Look at bottom graph in the first tweet

I prefer a super available test that always catches ppl w Ct <28 any day... over a barely available test w 3 day delay in results but that can catch to Ct 38

We shouldn't have even started looking for Cts of 32-40 for a virus like this.
Also - Thank you @jameshay218 for writing the script for me to make the plots :)
Here is a slightly different version
You can follow @michaelmina_lab.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: